paper-with-me

홈 › Papers

On the Identifiability of Quantized Factors

2023-06-28 · Vitória Barin-Pacela, Kartik Ahuja, Simon Lacoste-Julien, Pascal Vincent

Disentanglement aims to recover meaningful latent ground-truth factors from the observed distribution solely, and is formalized through the theory of identifiability. The identifiability of independent latent factors is proven to be impossible in the unsupervised i.i.d. setting under a general nonlinear map from factors to observations. In this work, however, we demonstrate that it is possible to recover quantized latent factors under a generic nonlinear diffeomorphism. We only assume that the latent factors have independent discontinuities in their density, without requiring the factors to be statistically independent. We introduce this novel form of identifiability, termed quantized factor identifiability, and provide a comprehensive proof of the recovery of the quantized factors.

📄 PDF Abstract BibTeX arXiv:2306.16334

Code (1)

facebookresearch/quantized_identifiability 공식 구현 pytorch

Tasks

DisentanglementInductive Bias

Similar Papers 제목 키워드 기반

Operationalizing Quantized Disentanglement

2025-11-25 · Vitoria Barin-Pacela, Kartik Ahuja, Simon Lacoste-Julien, Pascal Vincent arxiv

Recent theoretical work established the unsupervised identifiability of quantized factors under any diffeomorphism. The theory assumes that quantization thresholds correspond to axis-aligned discontinuities in the probab…

Multi-Channel Factor Analysis: Identifiability and Asymptotics

2024-07-26 · Gray Stanton, David Ramírez, Ignacio Santamaria, Louis Scharf 외

Recent work by Ram\'irez et al. [2] has introduced Multi-Channel Factor Analysis (MFA) as an extension of factor analysis to multi-channel data that allows for latent factors common to all channels as well as factors spe…

Leveraging Task Structures for Improved Identifiability in Neural Network Representations

2023-06-26 · Wenlin Chen, Julien Horwood, Juyeon Heo, José Miguel Hernández-Lobato

This work extends the theory of identifiability in supervised learning by considering the consequences of having access to a distribution of tasks. In such cases, we show that linear identifiability is achievable in the …

Representation Learning

Identifiability Results for Multimodal Contrastive Learning

2023-03-16 · Imant Daunhawer, Alice Bizeul, Emanuele Palumbo, Alexander Marx 외

Contrastive learning is a cornerstone underlying recent progress in multi-view and multimodal learning, e.g., in representation learning with image/caption pairs. While its effectiveness is not yet fully understood, a li…

Contrastive LearningRepresentation Learning

Identifiability in Two-Layer Sparse Matrix Factorization

2021-10-04 · Léon Zheng, Elisa Riccietti, Rémi Gribonval

Sparse matrix factorization is the problem of approximating a matrix $\mathbf{Z}$ by a product of $J$ sparse factors $\mathbf{X}^{(J)} \mathbf{X}^{(J-1)} \ldots \mathbf{X}^{(1)}$. This paper focuses on identifiability is…

Vocal Bursts Valence Prediction